Recent clinical prevention trials have demonstrated that chemoprevention of breast cancer is now possible. Results from the NSABP P1 trial showed that women at moderate risk of breast cancer who were treated with tamoxifen had a 45% reduction in breast cancer incidence. However, not all breast cancer was prevented, and the incidence of estrogen- receptor negative breast cancer was unchanged by tamoxifen treatment in this trial. Thus, there is an urgent need for more effective chemoprevention agents which could be used alone or combined with tamoxifen to prevent all forms of breast cancer. A major question which remains unanswered is whether women at very high risk of developing breast cancer, such as those with BRCA1 or BRCA2 mutations-those most in need of cancer risk reduction-will benefit from chemopreventive agents such as tamoxifen. In this project we propose a clinical trial to test a promising new chemopreventive agent, the RXR-selective retinoid Targretin, in a Phase II chemoprevention clinical trial. We hypothesize that this selective retinoid will suppress proliferation of breast tissue, and that women with BRCA1 and BRCA2 mutations will be sensitive to this anti-proliferative effect. In addition, we hypothesize that the RXR-selective retinoid will induce the expression of retinoid-regulated biomarkers associated with a non-malignant, differentiated phenotyped. To test these hypotheses, we will conduct a two arm, double-blinded chemoprevention trial, in which women at high risk of breast cancer will be treated with either Targretin or placebo for 1 month. These women will be screened for BRCA1 and BRCA2 mutations, and will have breast biopsies performed before and after the treatment so that changes in markers of proliferation, apoptosis, and differentiation state can be measured. Thus we will determine whether the RCR-selective retinoid inhibits proliferation and modulates marker expression, and whether the activity of the retinoid is affected by BRCA1 or BRCA2 mutational status. We will also investigate whether the treatment changes the expression of known retinoid-regulated markers (e.g. RARbeta, E-cadherin), or of newly identified markers found in Project 3 to be modulated by Targretin. The new markers will also be promising candidates for additional surrogate endpoints which could be tested in future Phase III clinical prevention trials. Changes in key markers in breast tissue from these women will support the pre-clinical evidence that the potentially less toxic RXR-selective retinoids can prevent cancer. These studies will provide critical information for designing and carrying out large-scale breast cancer prevention studies with these RXR-selective retinoids, and for selecting those populations most likely to respond.